Cross-linguistic generalizations about grammatical contexts favoring syncretism often have an implicational form. This paper shows that this is expected if (i) morphological paradigms are required to be both as small and as unambiguous as possible, (ii) languages may prioritize these requirements differently, and (iii) probability distributions for grammatical features interacting in syncretic patterns are fixed across languages. More specifically, this approach predicts that grammatical contexts that are less probable or more informative about a target grammatical feature $ T $ should favor syncretism of $ T $ cross-linguistically. The paper provides evidence for these predictions based on four detailed case studies involving well-known patterns of contextual syncretism (gender syncretism based on number, gender syncretism based on person, aspect syncretism based on tense, and case syncretism based on animacy).

In Tamil, coronals are licensed in onsets and initial syllables, exemplifying what Jesney (2011b) calls Licensing in Multiple Contexts (LMC). Jesney shows that while only positional faithfulness produces LMC in Optimality Theory, positional licensing provides a competing analysis of LMC in Harmonic Grammar (HG). This suggests that positional faithfulness may not be necessary in HG. We argue, though, that positional faithfulness remains essential. First, other facts in Tamil are incompatible with the positional licensing approach to LMC, rendering the positional faithfulness alternative the only viable analysis. Second, only with positional faithfulness can certain typological generalizations concerning assimilation between consonants be captured.

This study uses data from loanwords in Indonesian to argue for a phonological analysis using Harmonic Grammar (e.g. Smolensky & Legendre 2006, Pater, Bhatt & Potts 2007, Pater 2009). In original data consisting of Arabic and Dutch loanwords containing initial and final consonant clusters produced by 24 native speakers of Indonesian, we find both deletion and epenthesis to resolve word-final clusters, while word-initial clusters sometimes have epenthesis and sometimes are tolerated intact. The adaptations of Arabic and Dutch loanwords reveal the influence of three markedness constraints generally observed in Indonesian (*ComplexCoda, *ComplexOnset, and MinWord), and support a role for phonology in the analysis of borrowing, rather than a purely perceptual approach. When native monosyllables and borrowed monosyllables without clusters are considered, we find evidence that a standard Optimality Theory strict ranking is inadequate to account for the data; these constraints must be allowed to ‘gang up’, as in Harmonic Grammar, to account for the deletions, epenthesis, and non-adaptations found in the data.